*Official Benchmark Thread* - Post it here or it didn't happen :D

[Papusan]

16 minutes ago, electrosoft said:

 

I messaged them personally to tell them to not include it.... 😉

 

 

Why?😃 Because....

 

😁 😁 😁

[Mr. Fox]

OK, so I think this is going to be my daily driver overclock profile. Temps are OK without the chiller. That Sabrent kit was definitely better than the other generic kit. Runs 7200 with tight timings very stable.

 

---

I finally got around to testing the 2080 Ti with MATS and it has a failing RAM module (B0). I thoroughly cleaned and inspected the card and it looks like new otherwise, but I haven't done any probing. I am tempted to try a reflow on that one chip with my hot air station to see if it fixes it, but I probably should send it off for repair. It is still too good of a GPU to just kick it to the curb. Probably has many more years of service left if I fix it. @Rage SetI registered the card at EVGA and it is out of warranty. It has served me well and taken no prisoners in battle, too early for retirement.

mats version 400.184.  Testing TU102 with 20 MB of memory starting with 0 MB.

Read    Error Count: 0
Write   Error Count: 717805
Unknown Error Count: 0

=== MEMORY ERRORS BY SUBPARTITION ===
SUBPART READ ERRORS WRITE ERRORS UNKNOWN ERRS
------- ----------- ------------ ------------
FBIOA0            0            0            0
FBIOA1            0            0            0
FBIOB0            0       717805            0
FBIOB1            0            0            0
FBIOC0            0            0            0
FBIOC1            0            0            0
FBIOD0            0            0            0
FBIOD1            0            0            0
FBIOE0            0            0            0
FBIOE1            0            0            0
FBIOF0            0            0            0
FBIOF1            0            0            0

Failing Bits: 
   B000 B001 B002 B003 B004 B005 B006 B007 B008 B009 B010 B011 B012 B013 B014 B015 

=== MEMORY ERRORS BY BIT ===
P : Partition (FBIO)
                                            READ 0 READ 1 READ ?
P BIT READ ERRORS WRITE ERRORS UNKNOWN ERRS EXP. 1 EXP. 0 EXP. ?
- --- ----------- ------------ ------------ ------ ------ ------
B 000           0       776620            0 595624 180996      0
B 001           0       688440            0 271326 417114      0
B 002           0       806002            0 655095 150907      0
B 003           0       776706            0 180896 595810      0
B 004           0       688386            0 417028 271358      0
B 005           0       747347            0 210980 536367      0
B 006           0       688395            0 417049 271346      0
B 007           0       776663            0 180955 595708      0
B 008           0       629617            0 298088 331529      0
B 009           0       747307            0 211014 536293      0
B 010           0       776535            0 595472 181063      0
B 011           0       747350            0 210957 536393      0
B 012           0       776594            0 595601 180993      0
B 013           0       747319            0 211022 536297      0
B 014           0       688416            0 417131 271285      0
B 015           0       659160            0 301303 357857      0

 

 

report.txt

[Mr. Fox]

5 hours ago, tps3443 said:

With how advanced these new 13900K’s are. Using the newest Intel XTU is really the way to go with overclocking. It’s quite advanced! And user friendly. Plus, you are able to get peak potential from the cpu by leaving nothing on the table. 
 

I don’t use their auto tuning or anything. But manually overclocking with XTU is really awesome. It gives such great control over everything including the timings in real time. 
 

And I’m not having to jump in and out of the bios lol. 

Man, you're missing out using XTU brother. It's handy, yes, very, but it is a virus that consumes system resources and impairs performance. It adds extra services to your Windows installation.

 

You should be using Dragon Power and Dragon Ball for CPU and memory tuning. If the version I have works for 13th Gen you'll be set. Dragonball works on my ASUS Strix for memory tuning in Windows. Both programs work really well for me, especially Dragon Ball. Neither of them consumes resources like XTU does. When you apply settings and close them, there is nothing left running and neither one of them are modern UWP filth. They are "portable" apps that work on Windows 7.

MSI Dragon Ball1.0.0.08.zipMSI Dragon Power1.0.0.6-intel.zip

 

---

 

[Etern4l]

4 hours ago, ryan said:

@Etern4lI meant laptop as per what papusan said. going from 11800h to 12th gen is a pretty big jump...then 13th gen.....I bought this laptop in january of 2022 and its a 11800h and now we have 12th gen and 13th gen coming in. 

15 hours ago, Papusan said:

From what I remember he have 12800H aka 10nm chips (12th gen).

 

Cool, that makes more sense in terms of the 2 generations. Bro Papu mixed things up a bit this time 😉

Again, comparing laptop and desktop CPUs you lose out a lot within the same gen out of the gate, at least from 12th gen onwards. The idea of a "desktop replacement" laptop is ridiculously dead.

[jaybee83]

18 hours ago, tps3443 said:

My 13900KF overclocked is literally 300% faster than a stock 11900K in R23 MT. Pretty crazy. 
 

 

Great chips for sure. I’m testing mine at 5.5Ghz all cores, HT enabled, E-Cores enabled, and it’s peak power is 250 watts during R23 still scoring 40,660. I’m losing nothing in terms of performance and it brings the 13900K down near the AMD 7950X in terms of power levels. 

Good AIO could handle that. Definitely an improvement over last Gen.

 

those recent jumps in CPU performance are definitely insane. were approaching GPU style changes more and more, not just with performance jumps but also the way the hardware works (i.e. different core types, dynamic boosting behaviour, high number of cores, etc.)

just dont make the error of equaling such a golden chip with overall 13900K standards 😄 

 

17 hours ago, johnksss said:

So your card wont do 1700+ on memory?

 

well, i guess it depends what ure looking for. im testing for max. 24/7 stability on auto fan curve. for benching i think id be able to do +1650 on auto curve, at max fans and better case cooling 1700+ could be doable.

also, havent repasted / repadded anything yet, full stock.

 

10 hours ago, Papusan said:

NVIDIA GeForce RTX 4090 OC Shootout: ASUS, COLORFUL, FE, MSI, GAINWARD tweaktown.com
The ultimate GeForce RTX 4090 overclocking showdown: ASUS vs COLORFUL vs GAINWARD vs MSI vs Founders Edition... which overclocks the best?

 

Best Overclocker
You'd probably think it was the ASUS ROG Strix GeForce RTX 4090 OC Edition that is the best overclocker, but you'd be wrong... the ROG Strix RTX 4090 was topping out at just 2940MHz but your results may vary. The best overclocker that I've got here with my pile of custom GeForce RTX 4090 graphics cards is NVIDIA's own GeForce RTX 4090 Founders Edition which hit 3080MHz (but crashed) but was perfectly stable at 3060MHz.

 

Fun seeing Nvidia eat up their partners🙂 A good beginning before we see next chapter "eat their own" (5x90 FE). Have control on chips and memory chips bin quality/supply chain, pay off. Give it one of the best price tag help as well. Nvidia vs AIB partners = 1-0

 

huh, looks like mine aint that bad, currently still tuning the core but im at +200 offset resulting in load clocks of 3060-3120 Mhz depending on load type and core temps. full stock it runs 2850 and 2925 with maxed power / voltage limits (without clock offset).

but lets be honest, a total spread of 4% going from worst to best really aint nothing to write home about. i wouldnt even call that "binning", might as well just be (un)luck of the draw. still, its funny to see the most expensive card coming in last here lulz

 

8 hours ago, johnksss said:

Wait? Why no trash card in the list? Are they being haters?😂

 

maybe theyll do a separate piece on "trash card" performance 😂

 

4 hours ago, Mr. Fox said:

Man, you're missing out using XTU brother. It's handy, yes, very, but it is a virus that consumes system resources and impairs performance. It adds extra services to your Windows installation.

 

You should be using Dragon Power and Dragon Ball for CPU and memory tuning. If the version I have works for 13th Gen you'll be set. Dragonball works on my ASUS Strix for memory tuning in Windows. Both programs work really well for me, especially Dragon Ball. Neither of them consumes resources like XTU does. When you apply settings and close them, there is nothing left running and neither one of them are modern UWP filth. They are "portable" apps that work on Windows 7.

MSI Dragon Ball1.0.0.08.zip 3.82 MB · 0 downloads MSI Dragon Power1.0.0.6-intel.zip 4.81 MB · 0 downloads

 

---

 

 

thanks again for the software pointers bro Fox, unfortunately MSI Dragon Ball doesnt play BALL (pun intended) with my 7950X 😄 

[Mr. Fox]

26 minutes ago, jaybee83 said:

thanks again for the software pointers bro Fox, unfortunately MSI Dragon Ball doesnt play BALL (pun intended) with my 7950X 😄 

Dragon Ball works on my Asus mobos (present and past) not EVGA (present or past) but only Intel platforms. Dragon Power is a proprietary MSI-only product.

[Rage Set]

6 hours ago, Mr. Fox said:

OK, so I think this is going to be my daily driver overclock profile. Temps are OK without the chiller. That Sabrent kit was definitely better than the other generic kit. Runs 7200 with tight timings very stable.

 

---

I finally got around to testing the 2080 Ti with MATS and it has a failing RAM module (B0). I thoroughly cleaned and inspecte dthe card and it looks like new otherwise, but I haven't done any probing. I am tempted to try a reflow on that one chip with my hot air station to see if it fixes it, but I probably should send it off for repair. It is still too good of a GPU to just kick it to the curb. Probably has many more years of service left if I fix it. @Rage SetI registered the card at EVGA and it is out of warranty. It has served me well and taken no prisoners in battle, too early for retirement.

mats version 400.184.  Testing TU102 with 20 MB of memory starting with 0 MB.

Read    Error Count: 0
Write   Error Count: 717805
Unknown Error Count: 0

=== MEMORY ERRORS BY SUBPARTITION ===
SUBPART READ ERRORS WRITE ERRORS UNKNOWN ERRS
------- ----------- ------------ ------------
FBIOA0            0            0            0
FBIOA1            0            0            0
FBIOB0            0       717805            0
FBIOB1            0            0            0
FBIOC0            0            0            0
FBIOC1            0            0            0
FBIOD0            0            0            0
FBIOD1            0            0            0
FBIOE0            0            0            0
FBIOE1            0            0            0
FBIOF0            0            0            0
FBIOF1            0            0            0

Failing Bits: 
   B000 B001 B002 B003 B004 B005 B006 B007 B008 B009 B010 B011 B012 B013 B014 B015 

=== MEMORY ERRORS BY BIT ===
P : Partition (FBIO)
                                            READ 0 READ 1 READ ?
P BIT READ ERRORS WRITE ERRORS UNKNOWN ERRS EXP. 1 EXP. 0 EXP. ?
- --- ----------- ------------ ------------ ------ ------ ------
B 000           0       776620            0 595624 180996      0
B 001           0       688440            0 271326 417114      0
B 002           0       806002            0 655095 150907      0
B 003           0       776706            0 180896 595810      0
B 004           0       688386            0 417028 271358      0
B 005           0       747347            0 210980 536367      0
B 006           0       688395            0 417049 271346      0
B 007           0       776663            0 180955 595708      0
B 008           0       629617            0 298088 331529      0
B 009           0       747307            0 211014 536293      0
B 010           0       776535            0 595472 181063      0
B 011           0       747350            0 210957 536393      0
B 012           0       776594            0 595601 180993      0
B 013           0       747319            0 211022 536297      0
B 014           0       688416            0 417131 271285      0
B 015           0       659160            0 301303 357857      0

 

 

report.txt 75 kB · 0 downloads

 

 

That sucks bro, I'm sorry. I don't normally get the extended warranty on these cards because I don't own many of them for that long and it wouldn't transfer to a new owner anyway (unless I know them well and send it for repair on my behalf). Hopefully you can get it fixed. 

[Mr. Fox]

18 minutes ago, Rage Set said:

 

 

That sucks bro, I'm sorry. I don't normally get the extended warranty on these cards because I don't own many of them for that long and it wouldn't transfer to a new owner anyway (unless I know them well and send it for repair on my behalf). Hopefully you can get it fixed. 

I appreciate the condolences, but there's nothing for you to be sorry about. You blessed me with a great GPU at a wonderful price and there hasn't been a moment of dissatisfaction with it. The only reason it's not still working  flawlessly is my fault. I accidentally knocked my tower over while cleaning my office. When I turned it back on it had this problem. I'm thinking (hoping) that the shock of it getting slammed with a heavy waterblock bolted to it simply fractured a tiny ball of solder, and hoping it will make easier to repair than something that failed due to fatigue or electrical malfunction.

[johnksss]

2 hours ago, jaybee83 said:

well, i guess it depends what ure looking for. im testing for max. 24/7 stability on auto fan curve. for benching i think id be able to do +1650 on auto curve, at max fans and better case cooling 1700+ could be doable.

also, havent repasted / repadded anything yet, full stock.

 

 

huh, looks like mine aint that bad, currently still tuning the core but im at +200 offset resulting in load clocks of 3060-3120 Mhz depending on load type and core temps. full stock it runs 2850 and 2925 with maxed power / voltage limits (without clock offset).

but lets be honest, a total spread of 4% going from worst to best really aint nothing to write home about. i wouldnt even call that "binning", might as well just be (un)luck of the draw. still, its funny to see the most expensive card coming in last here lulz

 

 

maybe theyll do a separate piece on "trash card" performance 😂

 

 

 

I guess I mean out of the box doing nothing but set clocks and see if it can do it. No extra cooling or mods or repasting or anything else. 1500 and start port royal, no crazy artifacts(you can get always with about 10 to 15 seconds in and stop if one sees nothing), then go 1600 and then 1700 and then 1800. Up till it gets funky then back down a bit until stable and then run full test. Usually takes about 5 to 10 minutes max to find out if the memory is good or not.

[johnksss]

1 hour ago, Mr. Fox said:

Dragon Ball works on my Asus mobos (present and past) not EVGA (present or past) but only Intel platforms. Dragon Power is a proprietary MSI-only product.

Dragon Ball Power does not work well with 13900K....

[tps3443]

1 hour ago, Mr. Fox said:

I appreciate the condolences, but there's nothing for you to be sorry about. You blessed me with a great GPU at a wonderful price and there hasn't been a moment of dissatisfaction with it. The only reason it's not still working  flawlessly is my fault. I accidentally knocked my tower over while cleaning my office. When I turned it back on it had this problem. I'm thinking (hoping) that the shock of it getting slammed with a heavy waterblock bolted to it simply fractured a tiny ball of solder, and hoping it will make easier to repair than something that failed due to fatigue or electrical malfunction.

You can try heating it up with a heat gun a few cycles, see if it comes back around. Or send it off for an electronics X-ray and repair. It’s fairly cheap to do so.

[Papusan]

Two new in a rove. The smoke and fire show continue with nvidia's new modern power adapter. New has to be better. Always. Also Microsoft think so. Nvidia wanted to cut costs with this new design. 1600$ for the graphics cards isn't enough to make big enough profits. Hence they was forced to cut costs.

MSI Gaming X Trio RTX 4090 adapter burned

 

A Friend Messaged Me to check on my 4090's connector

Btw. What you think about the airflow direction on the fans below the fried 4090? Why use the bottom fans as  exhaust fans? So he could smell the smoke faster?

[Etern4l]

13 minutes ago, Papusan said:

Two new in a rove. The smoke and fire show continue with nvidia's new modern power adapter. New has to be better. Always. Also Microsoft think so. Nvidia wanted to cut costs with this new design. 1600$ for the graphics cards isn't enough to make big enough profits. Hence they was forced to cut costs.

MSI Gaming X Trio RTX 4090 adapter burned

 

A Friend Messaged Me to check on my 4090's connector

 

Phew, as I hoped, it's just the Nvidia Frankenstein adapter though - probably best to avoid using it.

 

The horror has a face – NVIDIA’s hot 12VHPWR adapter for the GeForce RTX 4090 with a built-in breaking point

 

 

 

[electrosoft]

Took my game drive over to my buddy's house earlier today (the one I sold my 12900ks) as he picked up a Suprim 4090 air cooled model  a few days ago from MC. He's still on the 12900ks and has no desire to upgrade to 13th. He went there to get a new M.2 for more storage and saw they had a few in stock and just said, "Ehhhh....why not?" 😉

 

 

My monitor tops out at 144hz while his is 4k 240hz (Xeneon).

 

WoW FPS gain I get ~140-150sh in the good spots in SL uncapped (I usually keep display capped at 140fps in NVCP) but outdoor raid bosses in SL can sink my 3090ti to the low 50s / high 40s. 4090?  Same spots hitting 210+. Those outdoor bosses can still sink a 4090 to 65-80fps but nowhere does it sink below 60fps anywhere everything ultra, RT to max. While the Nvidia readout said 65fps-80fps in a full encounter, it still felt so silky smooth versus how it can get a bit chunky with the 3090ti.

 

Ardenweald the zone GPU slayer? 3090ti = ~75fps avg, 4090 = 120fps.

 

What I confirmed was none of those Ryzen dips which are just brutal. I hope they work that out sooner than later but seeing as they are still there from the 5000 series ugh....

 

Fallout76 went from 80-150 up to 110-240.

 

Locking the fps to ~58fps in NVCP to simulate my 60hz G-sync 4k 32" BenQ and watching the pull just plummet to 140w-250w just about everywhere was nice to see. Even Ardenweald and outdoor raid bosses it topped out around 300w whereas my KPE 3090ti will hit 425-450w no problem.

 

Coil whine was pretty monstrous......luck of the draw I suppose. If I had picked that card up it would have been going right back.

 

Yeah, that 4090 is a beast. It isn't so much the highs for me as the min/lows it really pumps up quite nicely. If I switched back to my G-sync compatible BenQ 4k 32" 60hz display I would basically get that full 60fps locked in everywhere and most times the 4090 would be pulling so low too.

 

 

[Mr. Fox]

2 hours ago, Papusan said:

Btw. What you think about the airflow direction on the fans below the fried 4090? Why use the bottom fans as  exhaust fans? So he could smell the smoke faster?

Maybe the person that built the system wanted to pull air away from the GPU rather than pushing cooler fresh air to it. Some people act like they really hate their expensive computer parts. but I think it is often a matter of ignorance or poor judgment.

[johnksss]

 

[Mr. Fox]

21 minutes ago, johnksss said:

 

I wonder if a 7600X is a defective CPU with a dead CCX that can't be sold as the part it was originally intended to be. Rather than throwing it away, they sell it as a cheaper, less powerful processor.

[johnksss]

2 minutes ago, Mr. Fox said:

I wonder if a 7600X is a defective CPU with a dead CCX that can't be sold as the part it was originally intended to be. Rather than throwing it away, they sell it as a cheaper, less powerful processor.

You got me there. I was more looking at the delid tool they made for it along with the direct die frame.

[Mr. Fox]

5 minutes ago, johnksss said:

You got me there. I was more looking at the delid tool they made for it along with the direct die frame.

Yeah, both of them are nice products. Especially the delid tool. Very well engineered.

 

I hope he makes a die frame for 12th and 13th Gen Intel. I would love to ditch the IHS. I really miss running bare die.

[jaybee83]

On 10/28/2022 at 9:46 AM, johnksss said:

I guess I mean out of the box doing nothing but set clocks and see if it can do it. No extra cooling or mods or repasting or anything else. 1500 and start port royal, no crazy artifacts(you can get always with about 10 to 15 seconds in and stop if one sees nothing), then go 1600 and then 1700 and then 1800. Up till it gets funky then back down a bit until stable and then run full test. Usually takes about 5 to 10 minutes max to find out if the memory is good or not.

 

yep, so specifically for port royal it tops out at roughly +1600-1625, at 1650 and above it starts artifacting / crashing to desktop. im using a suite of 11 stress tests to finetune for 24/7 stability, port royal is just the first step i take to get a quick idea of performance increase and rough stability range. so basically started with +1600 in port royal and dialed it down step by step to +1550 while going through the suite.

btw, just finished up with the core, +200 stable 🙂 but im gonna try for +205 haha. in any case, clocks end up in the range of 3060-3120 depending on load and core temps. not bad!

highest power draw on the gpu itself that i saw was 560W, with gpu rails maxing out at a whopping 675W! crazy stuff, but temps were always under control, i think the absolute hottest hotspot temp that i saw was in the low to mid 80s in furmark / kombustor. so no worries on that front! usually its more like in the 70s, with junction temp in the mid 60s to low 70s.

 

On 10/28/2022 at 2:04 PM, electrosoft said:

Took my game drive over to my buddy's house earlier today (the one I sold my 12900ks) as he picked up a Suprim 4090 air cooled model  a few days ago from MC. He's still on the 12900ks and has no desire to upgrade to 13th. He went there to get a new M.2 for more storage and saw they had a few in stock and just said, "Ehhhh....why not?" 😉

 

 

My monitor tops out at 144hz while his is 4k 240hz (Xeneon).

 

WoW FPS gain I get ~140-150sh in the good spots in SL uncapped (I usually keep display capped at 140fps in NVCP) but outdoor raid bosses in SL can sink my 3090ti to the low 50s / high 40s. 4090?  Same spots hitting 210+. Those outdoor bosses can still sink a 4090 to 65-80fps but nowhere does it sink below 60fps anywhere everything ultra, RT to max. While the Nvidia readout said 65fps-80fps in a full encounter, it still felt so silky smooth versus how it can get a bit chunky with the 3090ti.

 

Ardenweald the zone GPU slayer? 3090ti = ~75fps avg, 4090 = 120fps.

 

What I confirmed was none of those Ryzen dips which are just brutal. I hope they work that out sooner than later but seeing as they are still there from the 5000 series ugh....

 

Fallout76 went from 80-150 up to 110-240.

 

Locking the fps to ~58fps in NVCP to simulate my 60hz G-sync 4k 32" BenQ and watching the pull just plummet to 140w-250w just about everywhere was nice to see. Even Ardenweald and outdoor raid bosses it topped out around 300w whereas my KPE 3090ti will hit 425-450w no problem.

 

Coil whine was pretty monstrous......luck of the draw I suppose. If I had picked that card up it would have been going right back.

 

Yeah, that 4090 is a beast. It isn't so much the highs for me as the min/lows it really pumps up quite nicely. If I switched back to my G-sync compatible BenQ 4k 32" 60hz display I would basically get that full 60fps locked in everywhere and most times the 4090 would be pulling so low too.

 

 

 

yep, can confirm the coil whine on my suprim x, as well. although it doesnt bother me at all, i actually just tested for it after i got the card after i saw so many users on overclock.net whining about that 😄 glad im not as sensitive towards that haha, cant hear it at all while gaming or benching.

and yes, the 4090 is such a monster in so many ways! can go up insanely high with power consumtion if u push it, at the same time it can be soooo frugal in terms of wattage per fps if u trim it down via UV or fps limiter. while playing wolfenstein the new order i brought it down to 90W avg for 60 fps locked at 4K everything maxed lulz

i absolutely love this card! 🙂 

 

On 10/28/2022 at 3:14 PM, Mr. Fox said:

Maybe the person that built the system wanted to pull air away from the GPU rather than pushing cooler fresh air to it. Some people act like they really hate their expensive computer parts. but I think it is often a matter of ignorance or poor judgment.

 

my guess is looks. just "looks nice" to see the front of the fans when looking into the case. unfortunately, that goes against any logic in terms of airflow 😂

 

On 10/28/2022 at 3:54 PM, Mr. Fox said:

Yeah, both of them are nice products. Especially the delid tool. Very well engineered.

 

I hope he makes a die frame for 12th and 13th Gen Intel. I would love to ditch the IHS. I really miss running bare die.

 

keeping an eye on this. might come in handy later down the line 🙂 

[electrosoft]

1 hour ago, jaybee83 said:

yep, can confirm the coil whine on my suprim x, as well. although it doesnt bother me at all, i actually just tested for it after i got the card after i saw so many users on overclock.net whining about that 😄 glad im not as sensitive towards that haha, cant hear it at all while gaming or benching.

and yes, the 4090 is such a monster in so many ways! can go up insanely high with power consumtion if u push it, at the same time it can be soooo frugal in terms of wattage per fps if u trim it down via UV or fps limiter. while playing wolfenstein the new order i brought it down to 90W avg for 60 fps locked lullz

i absolutely love this card! 🙂 

 

 

Bad coil whine, especially when it changes pitch/frequency depending on on-screen action literally makes me nauseous. I wish I could just push through it but I can't. I could even hear my FE 3090 through headphones. That Suprim was definitely talking! 😝

 

And exactly, pushing it to see the highs but for me the magic was capping it at 140 and 60 to simulate my two displays and just seeing the power consumption and heat drop massively as that would be exactly how I would use it. Seeing WoW locked in at 60fps  everywhere with rock solid fps 4k for the first time ever was a beautiful thing to see. 🙂

 

Even the air cooled Suprim is a gorgeous piece of hardware and the fan noise level wasn't nearly as bad as I thought it was going to be under full load. If I could find a low/no coil whine one of those, that would definitely be a contender along with the Liquid.

 

I know the Dragonflight areas in the Xpac have been heavily optimized versus SL so it should run even better. It drops Nov 28th so we'll see how the 3090ti does out there at 5120x1440  144hz and 3840x2160 60hz.

[tps3443]

1 hour ago, electrosoft said:

 

Bad coil whine, especially when it changes pitch/frequency depending on on-screen action literally makes me nauseous. I wish I could just push through it but I can't. I could even hear my FE 3090 through headphones. That Suprim was definitely talking! 😝

 

And exactly, pushing it to see the highs but for me the magic was capping it at 140 and 60 to simulate my two displays and just seeing the power consumption and heat drop massively as that would be exactly how I would use it. Seeing WoW locked in at 60fps  everywhere with rock solid fps 4k for the first time ever was a beautiful thing to see. 🙂

 

Even the air cooled Suprim is a gorgeous piece of hardware and the fan noise level wasn't nearly as bad as I thought it was going to be under full load. If I could find a low/no coil whine one of those, that would definitely be a contender along with the Liquid.

 

I know the Dragonflight areas in the Xpac have been heavily optimized versus SL so it should run even better. It drops Nov 28th so we'll see how the 3090ti does out there at 5120x1440  144hz and 3840x2160 60hz.

 

I can’t stand coil wine sometimes, especially when it’s really bad. My EVGA 1300P+ which is a high quality Platinum rated PSU with that big C19 sized power cable, well it has started screaming to death while gaming lol. All core heavy CPU load can’t trigger it. But if I play some games she straight up screams with the best.  My 3090KP doesn’t make a peep surprisingly.   

[johnksss]

Sorry guys, but I can't seem to let it go.😂😂 Zotac "trash" card was doing very respectable numbers. And really good vram and was up to about 550W max.

[tps3443]

UPDATED!! EDITED!! 263 watts!!

13900K@ 1.260V in bios LLC 5
900Khz VRM Switching frequency 
CPU Aux Load line at LLC 5
5.8Ghz P-Cores
4.5Ghz E-Cores
5.1Ghz Cache
DDR5 @7200 C34

R23 30-minute stability 

263 WATTS! YES! 263 WATTS OF JUICE! (30 min R23)

🤩

Edit: This CPU is a super chip!! 
 

Introducing the 13900KFS!

[Papusan]

Nvidia's reputation is at stake. And this only a few weeks before AMD launch their new graphics cards. AMD could'nt get a better before Christmas gift than what Nvidia gave them this time.

If you have a melted 4090 connector in the US you can report it to the Consumer Product Safety Commission here.

saferproducts.gov/

 

 

Fellow Canadian RTX 4090 owners, remember if you experience any melting issues with your adapter/connector, please report it to Health Canada

 

health.canada.ca/en/hea...

 

 

And this is just amusing... Always be prepared if you game on newest Nvidia graphics. Always have your fire extinguishers right beside the "ticking bomb". What can goes wrong here?😁 Btw... Who will have to pay for filling up the fire extinguishers after the fire due unsafe Nvidia products? Nvidia or you? Hehe 

 

And some others like to live on the utter edge... Cardboard packaging is as you know flammable😁